CN113945329B - SF (sulfur hexafluoride) 6 Method and system for judging gas leakage defect - Google Patents
SF (sulfur hexafluoride) 6 Method and system for judging gas leakage defect Download PDFInfo
- Publication number
- CN113945329B CN113945329B CN202111196653.8A CN202111196653A CN113945329B CN 113945329 B CN113945329 B CN 113945329B CN 202111196653 A CN202111196653 A CN 202111196653A CN 113945329 B CN113945329 B CN 113945329B
- Authority
- CN
- China
- Prior art keywords
- data
- gas
- pressure
- temperature
- fitting
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000007547 defect Effects 0.000 title claims abstract description 54
- 238000000034 method Methods 0.000 title claims abstract description 50
- 229910018503 SF6 Inorganic materials 0.000 title claims description 5
- SFZCNBIFKDRMGX-UHFFFAOYSA-N sulfur hexafluoride Chemical compound FS(F)(F)(F)(F)F SFZCNBIFKDRMGX-UHFFFAOYSA-N 0.000 title claims description 5
- 229960000909 sulfur hexafluoride Drugs 0.000 title claims description 5
- 238000007781 pre-processing Methods 0.000 claims abstract description 24
- 230000002159 abnormal effect Effects 0.000 claims description 20
- 238000000605 extraction Methods 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 12
- 238000012423 maintenance Methods 0.000 abstract description 9
- 230000006870 function Effects 0.000 description 12
- 230000008569 process Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000012360 testing method Methods 0.000 description 5
- 230000009471 action Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000012706 support-vector machine Methods 0.000 description 3
- 230000033228 biological regulation Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000013210 evaluation model Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000009413 insulation Methods 0.000 description 2
- 238000011418 maintenance treatment Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 238000010791 quenching Methods 0.000 description 2
- 230000000171 quenching effect Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000010292 electrical insulation Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000011282 treatment Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M3/00—Investigating fluid-tightness of structures
- G01M3/02—Investigating fluid-tightness of structures by using fluid or vacuum
- G01M3/04—Investigating fluid-tightness of structures by using fluid or vacuum by detecting the presence of fluid at the leakage point
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Examining Or Testing Airtightness (AREA)
Abstract
The application discloses an SF 6 Method and system for judging gas leakage defect, in particular for collecting SF in switchgear gas chamber 6 Gas data; preprocessing the gas data to obtain effective gas data; extracting characteristic parameters of the effective gas data to obtain a fitting parameter set; and judging the fitting parameter set through the OCSVM model, determining whether the fitting parameter set is in a target area, and outputting air leakage alarm information if the fitting parameter set exceeds the target area. And after the operation and maintenance personnel obtain the air leakage alarm information, corresponding maintenance measures can be timely taken. Can realize SF 6 Early defect early warning of the air chamber effectively improves the operation and maintenance level of the equipment and ensures the reliable operation of the equipment.
Description
Technical Field
The present application relates to the field of power equipment technology, and more particularly to a SF for a switchgear 6 A method and a system for judging gas leakage defects.
Background
In the switchgear industry, SF 6 The gas is widely applied to the opening of circuit breakers, GIS and the like due to the good electrical insulation performance and excellent arc extinguishing performanceIn the off-device. SF as an insulating and quenching medium 6 The density of the gas is a critical factor affecting its insulation and quenching capabilities, so SF is required to ensure gas performance 6 The density is kept within the standard range. In practice, as the operation time increases, the air chamber leaks air due to the defects of equipment aging, sealing damage and the like, and SF 6 The gas insulation and arc extinguishing properties are deteriorated therewith, thereby affecting the switching performance of the switching device. Thus requiring SF 6 And (5) monitoring the gas density in real time and early warning the gas leakage defect. In actual monitoring, in view of the practical situations such as the measurability of gas parameters, the usage habits, and the regulations corresponding to safety regulations, gas pressure values are currently used to replace the density of gas.
Currently aimed at SF 6 There are two methods for diagnosing the air leakage fault of the air chamber.
Firstly, judging a threshold value, and judging whether an air leakage fault exists or not by comparing the measured pressure with a preset alarm pressure; or comparing the characteristic quantity (such as pressure change rate, pressure first-order difference and the like) of the gas pressure with a set alarm threshold value, so as to judge whether the gas leakage occurs. This approach does not take into account the effect of environmental factors on pressure, especially the effect of temperature on pressure. In the early defect stage, the pressure change caused by air leakage can be disturbed by environmental factors, so that the method is insensitive to the early air leakage defect and cannot give an alarm in advance.
Secondly, machine learning is carried out, the method needs to collect gas data under normal operation and fault states, training and optimizing are carried out on the collected sample data to obtain a state evaluation model, and SF is evaluated by using the evaluation model 6 And (5) judging whether air leakage occurs or not in the running state. This method requires the collection of a certain amount of standard sample data, including normal operation data and fault data. However, in the field of switchgear, there is less data accumulation, and most of the sample data is obtained by analog collection in a laboratory, so that it is difficult to obtain field sample data meeting the requirements, and thus the accuracy and the feasibility of such a method may be problematic.
From the above, it can be seen that in the switching device SF 6 Air leakage of air chamberIn the direction of fault diagnosis, a method sensitive to early gas leakage defects and high in field applicability needs to be developed to solve the problem of SF of a switching equipment operation field 6 And detecting the early air leakage defect of the air chamber.
Disclosure of Invention
In view of this, the present application provides an SF 6 Method and system for judging gas leakage defect in gas chamber of switchgear 6 The gas state is monitored, and the early gas leakage defect alarm of the gas chamber is realized.
In order to achieve the above object, the following solutions have been proposed:
SF (sulfur hexafluoride) 6 Method for judging gas leakage defect, and SF 6 The method for judging the gas leakage defect comprises the following steps:
in-air SF of real-time acquisition switch equipment 6 Gas data;
preprocessing the gas data to obtain effective gas data;
performing linear fitting on the effective gas data to obtain the SF 6 Fitting parameter set (k), b of gas data;
inputting the fitting parameter set into a trained OCSVM model, judging whether the fitting parameter set exceeds a target area, and judging that the air chamber is not air-leaking if the fitting parameter set is in the target area; otherwise, judging that the air chamber leaks air, and outputting air leakage alarm information.
Optionally, the gas data includes pressure data and temperature data.
Optionally, extracting characteristic parameters of the effective gas data to obtain the SF 6 Fitting parameter set (k, b) of gas data, comprising the steps of:
grouping the effective gas data by taking a day as a time unit, and performing pressure-temperature linear fitting on each group of the effective gas data to obtain a fitting slope k and a fitting intercept b, thereby forming the fitting parameter group (k, b).
Optionally, the ocvm model is obtained through the following steps:
collecting normal operation pressure-temperature data of the switching equipment air chamber;
preprocessing the normal operation pressure-temperature data to obtain effective pressure-temperature data;
extracting characteristic quantities aiming at the effective pressure-temperature data to obtain a fitting parameter group (k, b);
and constructing an initial model, and inputting the effective pressure-temperature data into the initial model to perform model training to obtain the OCSVM model.
Optionally, the method further comprises the steps of:
SF-based 6 And (3) carrying out pressure trend prediction by the gas temperature prediction model and the pressure-temperature relation fitting parameter group (k, b).
SF of switch equipment 6 Gas leakage defect research and judgment system, SF 6 The gas leakage defect judging system includes:
the data acquisition module is used for acquiring SF in the switchgear air chamber in real time 6 Gas data;
the data preprocessing module is used for preprocessing the gas data to obtain effective gas data;
the parameter extraction module is used for extracting characteristic parameters of the effective gas data to obtain the SF 6 Fitting parameter sets of gas data;
the alarm output module is used for inputting the fitting parameter set into the trained OCSVM model, judging whether the fitting parameter set is in a target area, and judging that the air chamber is not air-leaking if the fitting parameter set is in the target area; otherwise, judging that the air chamber leaks air, and outputting air leakage alarm information.
Optionally, the gas data includes pressure data and temperature data.
Optionally, the parameter extraction module includes:
a data grouping unit for grouping data in a unit of time of day;
fitting processing unit for each group of SF 6 And (3) carrying out linear fitting on the gas pressure and temperature data to obtain a fitting slope k and a fitting intercept b, and forming the fitting parameter group (k, b).
Optionally, the method further comprises:
the acquisition execution module is used for acquiring the normal operation pressure-temperature data of the switching equipment air chamber;
the processing execution module is used for preprocessing the normal operation pressure-temperature data to obtain effective pressure-temperature data;
the extraction execution module is used for extracting characteristic quantities aiming at the effective pressure-temperature data to obtain a fitting parameter group (k, b);
and the model training module is used for constructing an initial model, inputting the effective pressure-temperature data into the initial model for model training, and obtaining the OCSVM model.
Optionally, the method further comprises:
trend prediction module for passing SF 6 And (3) carrying out pressure trend prediction by the gas temperature prediction model and the pressure-temperature relation fitting parameter group (k, b).
As can be seen from the technical scheme, the application discloses an SF 6 Method and system for judging gas leakage defect, in particular for collecting SF in switchgear gas chamber 6 Gas data; preprocessing the gas data to obtain effective gas data; extracting characteristic parameters of the effective gas data to obtain a fitting parameter set; and judging the fitting parameter set through the OCSVM model, determining whether the fitting parameter set is in a target area, and outputting air leakage alarm information if the fitting parameter set is not in the target area. And after the operation and maintenance personnel obtain the air leakage alarm information, corresponding treatment measures can be timely taken. The method can realize SF 6 Early defect early warning of the air chamber effectively improves the operation and maintenance level of the equipment and ensures the reliable operation of the equipment.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 shows an SF according to an embodiment of the present application 6 A flow chart of a gas leakage defect judging method;
FIG. 2a shows SF in an embodiment of the present application 6 A gas state graph;
FIG. 2b shows SF in an embodiment of the present application 6 A pressure-temperature fit of the gas;
FIG. 2c is a schematic diagram of an OCSVM model according to an embodiment of the present application;
FIG. 3 is another SF of an embodiment of the present application 6 A flow chart of a gas leakage defect judging method;
FIG. 4 shows SF of an embodiment of the present application 6 A pressure trend prediction graph of the gas;
fig. 5 shows an SF according to an embodiment of the present application 6 A block diagram of a gas leakage defect judging system;
FIG. 6 is another SF of an embodiment of the present application 6 A block diagram of a gas leakage defect judging system;
FIG. 7 is a further SF of an embodiment of the present application 6 And a block diagram of a gas leakage defect judging system.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.
Example 1
Fig. 1 shows an SF according to an embodiment of the present application 6 A flow chart of a method for judging gas leakage defects.
As shown in fig. 1, the SF provided in this embodiment 6 The gas leakage defect judging method is used for real-time evaluating the operation state of the gas chamber of the switch equipment to determine whether the gas chamber has a gas leakage defect or not, and the SF 6 The method for judging the gas leakage defect comprises the following steps:
s1, collecting SF in switchgear air chamber in real time 6 Gas data.
The gas data comprise temperature data and pressure data, the corresponding sensors are communicated with the gas chamber, and the specific positions can be flexibly selected according to requirements. When the switch equipment is assembled and debugged on site and runs stably, the temperature data and the pressure data are collected according to a preset sampling frequency so as to carry out subsequent processing.
S2, preprocessing the acquired gas data to obtain effective gas data.
The collected gas data belongs to original data, and abnormal values and missing values may exist, so that the original data needs to be preprocessed, and the method specifically comprises the following steps:
the abnormal values in the gas data include format abnormal data and value abnormal data. Aiming at the format abnormal data, directly deleting the format abnormal data; and judging whether the value abnormal data belongs to error data or not, if so, performing deleting operation, and if not, reserving the data.
There are two methods for missing value handling: one is to delete the array containing missing value directly, this method is suitable for the situation that the data size is larger, the missing value proportion is smaller; the other is missing value filling, and the missing values are filled by using statistical values, wherein common filling values comprise mean values, median values, mode values and the like. In the actual processing, a proper processing method is selected according to the specific condition of the missing value.
And S3, extracting characteristic parameters of the effective gas data to obtain a fitting parameter set.
The SF is obtained by extracting the characteristic parameters of the obtained effective gas data 6 Characteristic parameter sets (k, b) of the gas data.
According to the experimental rule of the gas, the ideal gas meets the ideal gas state equation. SF (sulfur hexafluoride) 6 The molecular mass of the gas is large, and the intermolecular acting force is obvious. When the absolute pressure is more than 0.3MPa, the intermolecular attraction force is enhanced with the increase of the gas density and the decrease of the intermolecular distance, and the gasThe pressure will no longer conform to the ideal gas equation. At this time, the Beattie-Bridgman equation is used to describe SF 6 The gas state parameter relationship is shown in the following formula 1:
P＝(56.9×ρ×T×(1+B)-B 2 ×A)×10 -6 (1)
wherein:
A＝74×(1-0.727×10 -3 ×ρ)
B＝2.51×10 -3 ×ρ×(1-0.846×10 -3 ×ρ)
wherein:
P—SF 6 gas pressure (gauge pressure), MPa;
ρ—SF 6 density of gas, kg/m 3 ；
T—SF 6 Absolute temperature of gas, K.
As can be seen from the above, SF 6 The pressure is determined by the gas temperature and the gas density, when SF 6 When the gas density is unchanged, the gas pressure and the temperature are in linear relation, thereby obtaining SF 6 The gas state curve is shown in fig. 2 a.
SF-based 6 The gas state parameter relationship and the gas state curve can show that for a certain gas density, the gas pressure and the temperature are in a linear relationship, and the linear relationship related parameters, namely the slope (k) and the intercept (b), are related to the gas density, namely the slope and the intercept of the gas pressure and the temperature can reflect SF 6 The density is varied, and these two parameters are only related to the gas density and are not affected by temperature. The application therefore groups the acquired data in time units of days for each group of SF 6 The gas pressure was linearly fitted to the temperature data as shown in fig. 2 b. From the above fitting procedure, a set of fitting parameters (k, b) is obtained, which set of parameters can characterize SF 6 The gas density varies. Where k refers to the pressure-temperature linear fit slope parameter and b refers to the pressure-temperature linear fit intercept parameter.
And S4, judging the fitting parameter set through the OCSVM model, and outputting air leakage early warning information.
Specifically, performing linear fitting on the gas pressure-temperature data of each day to obtain a fitting parameter set (k, b), inputting the fitting parameter set into a trained OCSVM model, judging whether the fitting parameter set is in a target area, and if so, enabling the gas chamber to be in a normal running state at the moment; if the fitting parameter set is not in the target area, the gas leakage defect of the gas chamber at the moment can be judged, and alarm information is sent out. In addition, the corresponding fitting parameter set is updated and output while the air leakage defect alarm is sent out.
The OCSVM (one class support vector machine) model, i.e. a single class support vector machine, maps data samples to a high-dimensional feature space through kernel functions, making them more aggregated. The object solves an optimal hyperplane in the feature space to separate object data from the origin of coordinates, and maximizes the distance from the separation hyperplane to the zero point. This process produces a binary function that can capture the probability density region of the data in the feature space. When the data point is in this region, return +1, return-1 in the other regions. The model is specifically shown in fig. 2c, where the gray area is a probability density area obtained by training, that is, the target area mentioned in the application, if the test data point is in the target area, the test data point returns to +1, otherwise returns to-1.
The optimization objective of the OCSVM model is shown in formula (2): :
wherein: phi (x) i ) Representing a kernel function ζ i Representing the relaxation variable, v is the degree of penalty for classifying errors in the case of linear inseparability.
After determining ω, ρ by calculation, the model function becomes:
f(x)＝sgn((ω T φ(x i ))-ρ) (3)
the function is a hyperplane of parameter ω, ρ that is the greatest distance from the zero point of the feature space and separates the zero point from all data points. Let variable x i And inputting the function and returning a classification result.
Specifically, the OCSVM model is obtained by the following steps:
firstly, collecting pressure-temperature data of normal operation of a gas chamber of the switching equipment in a certain time, namely internal pressure and internal temperature corresponding to the certain time.
Then, the normal operation pressure-temperature data is preprocessed, wherein the preprocessing is the same as the preprocessing in the step S2, and is not described herein, so as to obtain effective pressure-temperature data.
And then, extracting the characteristic quantity of the effective pressure-temperature data to obtain a fitting parameter group (k, b), wherein the extraction process is the same as that in the step S3, and the description is omitted.
And finally, constructing an initial model, inputting effective pressure-temperature data into the initial model for model training, and obtaining the OCSVM model.
As can be seen from the above technical solutions, the present embodiment provides an SF 6 Method for judging gas leakage defect, in particular for collecting SF in gas chamber of switch equipment 6 Gas data; preprocessing the gas data to obtain effective gas data; extracting characteristic parameters of the effective gas data to obtain SF 6 Fitting parameter sets (k, b) of the gas data; and judging the fitting parameter set based on the OCSVM model, determining whether the fitting parameter set is in a target area, and outputting air leakage defect alarm information if the fitting parameter set (k, b) is not in the target area. And after the air leakage alarm information is obtained, the operation and maintenance personnel can take corresponding operation and maintenance treatment measures in time. The method can realize SF 6 Early defect early warning of the air chamber effectively improves the operation and maintenance level of the equipment and ensures the reliable operation of the equipment.
The embodiment determines SF by a mechanism-based data analysis method 6 The relationship between gas pressure and temperature is obtained by linear fitting of pressure-temperature data to obtain a set of fitting parameters (k, b) which are related only to density, and SF can be measured 6 Density and can reflect SF more directly 6 A gas leakage condition; the OCSVM is adopted to classify the fitting parameter group, defect sample data is not needed, and SF can be realized 6 Early defect of air chamberEarly warning and SF improvement 6 The accuracy and timeliness of gas leakage defect early warning.
In addition, in a specific embodiment of the present application, the method further includes the following steps, as shown in fig. 3:
s5, SF in the opposite air chamber 6 The gas pressure is trend predicted.
As can be seen from step S3, SF 6 The gas pressure and the temperature have linear correlation, and the temperature has stronger stability and periodicity relative to the pressure, so the gas pressure has higher predictability, therefore, in the patent, an LSTM (Long Short-Term Memory artificial neural network) algorithm is adopted as a temperature prediction model to conduct time series prediction on the temperature, and on the basis of the temperature prediction, the trend prediction is conducted on the gas pressure by combining the fitting parameter set (k, b) obtained in the step S3, and the prediction result is shown in a figure 4. As can be seen from fig. 4, SF 6 The prediction error of the pressure trend is smaller, and the prediction accuracy requirement can be met.
At SF 6 Among relevant parameters, the temperature has stronger stability and periodicity, so the temperature has higher predictability, and the accuracy of pressure trend prediction can be improved by carrying out pressure trend prediction through the fitting parameter group (k, b) updated in real time by the temperature.
Example two
Fig. 5 shows an SF according to an embodiment of the present application 6 And a block diagram of a gas leakage defect judging system.
As shown in fig. 5, the SF provided in this embodiment 6 The gas leakage defect judging system is used for real-time evaluating the operation state of the gas chamber of the switchgear to determine whether the gas leakage defect exists or not, and the SF 6 The gas prediction system comprises a data acquisition module 10, a data preprocessing module 20, a parameter extraction module 30 and an alarm output module 40.
The SF 6 The method for judging the gas leakage defect comprises the following steps:
the data acquisition module is used for acquiring SF in the switchgear air chamber in real time 6 Gas data.
The gas data comprise temperature data and pressure data, the corresponding sensors are communicated with the gas chamber, and the specific positions can be flexibly selected according to requirements. When the switch equipment is assembled and debugged on site and runs stably, the temperature data and the pressure data are collected according to a preset sampling frequency so as to carry out subsequent processing.
The data preprocessing module is used for preprocessing the gas data to obtain effective gas data.
The collected gas data belongs to original data, and abnormal values and missing values may exist, so that the original data needs to be preprocessed, and the method specifically comprises the following steps:
the abnormal value in the gas data includes format abnormal data and value abnormal data. Aiming at the format abnormal data, directly deleting the format abnormal data; and judging whether the value abnormal data belongs to error data or not, if so, performing deleting operation, and if not, reserving the data.
There are two methods for missing value handling: one is to delete the array containing missing value directly, this method is suitable for the situation that the data size is larger, the missing value proportion is smaller; the other is missing value filling, and the missing values are filled by using statistical values, wherein common filling values comprise mean values, median values, mode values and the like. In the actual processing, a proper processing method is selected according to the specific condition of the missing value.
The parameter extraction module is used for extracting characteristic parameters of the effective gas data to obtain a fitting parameter group (k, b).
The SF is obtained by extracting the characteristic parameters of the obtained effective gas data 6 Fitting parameter sets (k, b) of the gas data.
According to the experimental rule of the gas, the ideal gas meets the ideal gas state equation and SF 6 The molecular mass of the gas is large, and the intermolecular acting force is obvious. When the absolute pressure is greater than 0.3MPa, the intermolecular attraction force is enhanced with the increase of the gas density and the decrease of the intermolecular distance, and the pressure of the gas no longer accords with the ideal gas equation. The curve processing unit is used for describing SF by adopting Beattie-Bridgman formula 6 Gaseous state parameterThe numerical relationship is shown in the following formula 4:
P＝(56.9×ρ×T×(1+B)-B 2 ×A)×10 -6 (4)
wherein:
A＝74×(1-0.727×10 -3 ×ρ)
B＝2.51×10 -3 ×ρ×(1-0.846×10 -3 ×ρ)
wherein:
P—SF 6 gas pressure (gauge pressure), MPa;
ρ—SF 6 density of gas, kg/m 3 ；
T—SF 6 Absolute temperature of gas, K.
As can be seen from the above, SF 6 The pressure is determined by the gas temperature and the gas density, when SF 6 Determining the density of the gas, and obtaining SF by linearly correlating the pressure of the gas with the temperature 6 The gas state curve is shown in fig. 2a below.
SF-based 6 The gas state parameter relationship and the gas state curve show that the gas pressure and the temperature are in a linear relationship for a fixed gas density, and the linear relationship related parameters, namely the slope (k) and the intercept (b), are related to the gas density, namely the slope and the intercept of the gas pressure and the temperature can reflect SF 6 The density is varied, and these two parameters are only related to the gas density and are not affected by temperature.
Fitting processing unit for SF for daily monitoring 6 The gas pressure-temperature data were linearly fitted as shown in fig. 2 b. From the above fitting procedure, a set of fitting parameters (k, b) is obtained, which set of parameters can characterize SF 6 The gas density varies. Where k refers to the pressure-temperature linear fit slope parameter and b refers to the pressure-temperature linear fit intercept parameter.
And the alarm output module is used for judging the fitting parameter set through the OCSVM model and outputting air leakage early warning information.
Specifically, for each day of fitting parameter set (k, b) of gas data, inputting the fitting parameter set into a trained OCSVM model, judging whether the fitting parameter set is in a target area, and if so, judging that the gas chamber is in a normal running state at the moment; if the fitting parameter set is not in the target area, the gas leakage defect of the gas chamber at the moment can be judged, and alarm information is sent out. In addition, the corresponding fitting parameter set is updated and output while the air leakage defect alarm is sent out.
The OCSVM model is a single-class support vector machine, and maps data samples to a high-dimensional feature space through a kernel function, so that the model has better aggregation. The object solves an optimal hyperplane in the feature space to separate object data from the origin of coordinates, and maximizes the distance from the separation hyperplane to the zero point. This process produces a binary function that can capture the probability density region of the data in the feature space. When the data point is in this region, return +1, return-1 in the other regions. The model is specifically shown in fig. 2c, wherein the gray area in the figure is a probability density area obtained by training, if the test data point is in the gray area, the test data point returns to +1, otherwise, the test data point returns to-1.
The optimization objective of the OCSVM model is shown in formula (2): :
wherein: phi (x) i ) Representing a kernel function ζ i Representing the relaxation variable, v is the degree of penalty for classifying errors in the case of linear inseparability.
After determining ω, ρ by calculation, the model function becomes:
f(x)＝sgn((ω T φ(x i ))-ρ) (3)
the function is a hyperplane of parameter ω, ρ that is the greatest distance from the zero point of the feature space and separates the zero point from all data points. Let variable x i And inputting the function and returning a classification result.
In addition, in a specific implementation of this embodiment, the system further includes an acquisition execution module 50, a processing execution module 60, an extraction execution module 70, and a model training module 80, as shown in fig. 6.
The acquisition execution module is used for acquiring pressure-temperature data of the normal operation of the switching equipment air chamber in a certain time, namely internal pressure and internal temperature corresponding to a certain time.
The processing execution module is configured to perform preprocessing on the normal operating pressure-temperature data, where the preprocessing is the same as the preprocessing in step S2, and is not described herein again, so as to obtain effective pressure-temperature data.
The extraction execution module is used for extracting characteristic quantities aiming at the effective pressure-temperature data to obtain a fitting parameter group (k, b); the extraction process is the same as the process in step S3, and will not be described here again.
The model training module is used for inputting the effective pressure-temperature data into the initial model to perform model training, and the OCSVM model is obtained.
As can be seen from the above technical solutions, the present embodiment provides an SF for a switchgear 6 Gas leakage defect research and judgment system, in particular to gas chamber SF of acquisition switch equipment 6 Gas data; preprocessing the gas data to obtain effective gas data; extracting characteristic parameters of the effective gas data to obtain SF 6 Fitting parameter sets (k, b) of the gas data; and judging the fitting parameter set based on the OCSVM model, determining whether the fitting parameter set is in the target area, and outputting air leakage defect alarm information if the fitting parameter set is not in the target area. And after the air leakage alarm information is obtained, the operation and maintenance personnel can take corresponding operation and maintenance treatment measures in time. The method can realize SF 6 Early defect early warning of the air chamber effectively improves the operation and maintenance level of the equipment and ensures the reliable operation of the equipment.
The embodiment determines SF by a mechanism-based data analysis method 6 The relationship between gas pressure and temperature is obtained by linear fitting of pressure-temperature data to obtain a set of fitting parameters (k, b) which are related only to density, and SF can be measured 6 Density and can reflect SF more directly 6 A gas leakage condition; the OCSVM is adopted to classify the fitting parameter group, defect sample data is not needed, and SF can be realized 6 Early defect early warning of air chamber and SF improvement 6 The accuracy and timeliness of gas leakage defect early warning.
In addition, in a specific embodiment of the present application, a trend prediction module 90 is further included, as shown in fig. 7:
the trend prediction module is used for predicting SF in the air chamber 6 The gas pressure is trend predicted.
SF is known by a parameter extraction module 6 The gas pressure and the temperature have stronger linear relation, and the temperature has stronger stability and periodicity relative to the pressure, so the temperature has higher predictability, therefore, the LSTM algorithm is adopted as a temperature prediction model in the patent to predict the temperature in time series, and on the basis of temperature prediction, the fitting parameter group (k, b) obtained by the parameter extraction module is combined to predict the trend of the gas pressure, and the prediction result is shown in figure 4.
As can be seen from fig. 4, SF 6 The prediction error of the pressure trend is smaller, and the prediction accuracy requirement can be met.
At SF 6 Among relevant parameters, the temperature has stronger stability and periodicity, so the temperature has higher predictability, and the accuracy of pressure trend prediction can be improved by carrying out pressure trend prediction through the fitting parameter group (k, b) updated in real time by the temperature.
In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described by differences from other embodiments, and identical and similar parts between the embodiments are all enough to be referred to each other.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the scope of the embodiments of the invention.
Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or terminal device comprising the element.
The foregoing has outlined rather broadly the more detailed description of the invention in order that the detailed description of the invention that follows may be better understood, and in order that the present principles and embodiments may be better understood; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in accordance with the ideas of the present invention, the present description should not be construed as limiting the present invention in view of the above.
Claims (4)
1. SF (sulfur hexafluoride) 6 The method for judging the gas leakage defect is characterized in that 6 The method for judging the gas leakage defect comprises the following steps:
in-air SF of real-time acquisition switch equipment 6 Gas data;
preprocessing the gas data to obtain effective gas data, including: aiming at the format abnormal data, directly deleting the format abnormal data; judging whether the value abnormal data belong to error data or not, if so, performing deleting operation, and if not, reserving the data; for the missing values, directly deleting the array containing the missing values, or filling the missing values with statistical values, wherein the statistical values at least comprise a mean value, a median value and a mode value; the gas data includes pressure data and temperature data;
performing linear fitting on the effective gas data to obtain the SF 6 Fitting parameter sets (k, b) of the gas data;
inputting the fitting parameter set into a trained OCSVM model, judging whether the fitting parameter set exceeds a target area, and judging that the air chamber is not air-leaking if the fitting parameter set is in the target area; otherwise, judging that the air chamber leaks air, and outputting leakage alarm information;
the effective gas data is subjected to linear fitting to obtain the SF 6 Fitting parameter set (k, b) of gas data, comprising the steps of:
grouping the effective gas data by taking a day as a time unit, and performing pressure-temperature linear fitting on each group of the effective gas data to obtain a fitting slope k and a fitting intercept b, thereby forming the fitting parameter group (k, b);
the OCSVM model is obtained through the following steps:
collecting normal operation pressure-temperature data in an air chamber of the switching equipment;
preprocessing the normal operation pressure-temperature data to obtain effective pressure-temperature data;
extracting characteristic quantities aiming at the effective pressure-temperature data to obtain a fitting parameter group (k, b);
and constructing an initial model, and inputting the effective pressure-temperature data into the initial model to perform model training to obtain the OCSVM model.
2. The SF of claim 1 6 The method for judging the gas leakage defect is characterized by further comprising the following steps:
SF-based 6 And (3) carrying out pressure trend prediction by the gas temperature prediction model and the pressure-temperature relation fitting parameter group (k, b).
3. SF of switch equipment 6 The system for judging the gas leakage defect is characterized in that 6 The gas leakage defect judging system includes:
the data acquisition module is used for acquiring SF in the switchgear air chamber in real time 6 Gas data;
the data preprocessing module is used for preprocessing the gas data to obtain effective gas data; the gas data includes pressure data and temperature data;
the parameter extraction module is used for extracting characteristic parameters of the effective gas data to obtain the SF 6 Fitting parameter sets of gas data;
the alarm output module is used for inputting the fitting parameter set into the trained OCSVM model, judging whether the fitting parameter set is in a target area, and judging that the air chamber is not air-leaking if the fitting parameter set is in the target area; otherwise, judging that the air chamber leaks air, and outputting leakage alarm information;
the data preprocessing module is specifically used for: aiming at the format abnormal data, directly deleting the format abnormal data; judging whether the value abnormal data belong to error data or not, if so, performing deleting operation, and if not, reserving the data; for the missing values, directly deleting the array containing the missing values, or filling the missing values with statistical values, wherein the statistical values at least comprise a mean value, a median value and a mode value;
the parameter extraction module comprises:
data packet unit for SF in days as time unit 6 Grouping gas data;
fitting processing unit for each group of SF 6 Performing linear fitting on the gas pressure and temperature data to obtain a fitting slope k and a fitting intercept b, and forming the fitting parameter group (k, b);
further comprises:
the acquisition execution module is used for acquiring normal operation pressure-temperature data in the air chamber of the switching equipment;
the processing execution module is used for preprocessing the normal operation pressure-temperature data to obtain effective pressure-temperature data;
the extraction execution module is used for extracting characteristic quantities aiming at the effective pressure-temperature data to obtain a fitting parameter group (k, b);
and the model training module is used for constructing an initial model, inputting the effective pressure-temperature data into the initial model for model training, and obtaining the OCSVM model.
4. A SF according to claim 3 6 The gas leakage defect judging system is characterized by further comprising:
trend prediction module for passing SF 6 And (3) carrying out pressure trend prediction by the gas temperature prediction model and the pressure-temperature relation fitting parameter group (k, b).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111196653.8A CN113945329B (en) | 2021-10-14 | 2021-10-14 | SF (sulfur hexafluoride) 6 Method and system for judging gas leakage defect |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111196653.8A CN113945329B (en) | 2021-10-14 | 2021-10-14 | SF (sulfur hexafluoride) 6 Method and system for judging gas leakage defect |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113945329A CN113945329A (en) | 2022-01-18 |
| CN113945329B true CN113945329B (en) | 2024-01-23 |
Family
ID=79329838
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111196653.8A Active CN113945329B (en) | 2021-10-14 | 2021-10-14 | SF (sulfur hexafluoride) 6 Method and system for judging gas leakage defect |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113945329B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH11296766A (en) * | 1998-04-09 | 1999-10-29 | Yazaki Corp | Gas leakage alarm and gas leakage alarming system |
| KR20080085390A (en) * | 2007-03-19 | 2008-09-24 | 주식회사 싸이언 | Measuring appratus for leaking gas and vacuum leak using sf6 gas |
| WO2015149597A1 (en) * | 2014-04-04 | 2015-10-08 | 袁志贤 | Iec61850-based communication simulation method for sf6 gas on-line monitoring device |
| CN107024672A (en) * | 2017-04-26 | 2017-08-08 | 贵州电网有限责任公司电力科学研究院 | A kind of SF6The operation scaling method of insulation of electrical installation state on_line monitoring system |
| CN107328518A (en) * | 2017-07-13 | 2017-11-07 | 国网浙江省电力公司检修分公司 | GIS device SF6 on-line monitoring systems and its method of work |
| CN108020781A (en) * | 2017-12-19 | 2018-05-11 | 上海电机学院 | A kind of circuit breaker failure diagnostic method |
| CN109190828A (en) * | 2018-09-07 | 2019-01-11 | 苏州大学 | Gas leakage concentration distribution determines method, apparatus, equipment and readable storage medium storing program for executing |
| CN110689324A (en) * | 2019-09-29 | 2020-01-14 | 国网湖北省电力有限公司电力科学研究院 | Auxiliary judging method for detection result of distribution transformer |
| CN111612128A (en) * | 2020-05-14 | 2020-09-01 | 国网江苏省电力有限公司电力科学研究院 | Active early warning method and device for SF6 high-voltage circuit breaker state |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7257496B2 (en) * | 2005-07-28 | 2007-08-14 | Avistar, Inc. | Method and apparatus for monitoring SF6 gas and electric utility apparatus |
| JP5171135B2 (en) * | 2007-07-09 | 2013-03-27 | 東光東芝メーターシステムズ株式会社 | Gas leak detection device and method |
-
2021
- 2021-10-14 CN CN202111196653.8A patent/CN113945329B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH11296766A (en) * | 1998-04-09 | 1999-10-29 | Yazaki Corp | Gas leakage alarm and gas leakage alarming system |
| KR20080085390A (en) * | 2007-03-19 | 2008-09-24 | 주식회사 싸이언 | Measuring appratus for leaking gas and vacuum leak using sf6 gas |
| WO2015149597A1 (en) * | 2014-04-04 | 2015-10-08 | 袁志贤 | Iec61850-based communication simulation method for sf6 gas on-line monitoring device |
| CN107024672A (en) * | 2017-04-26 | 2017-08-08 | 贵州电网有限责任公司电力科学研究院 | A kind of SF6The operation scaling method of insulation of electrical installation state on_line monitoring system |
| CN107328518A (en) * | 2017-07-13 | 2017-11-07 | 国网浙江省电力公司检修分公司 | GIS device SF6 on-line monitoring systems and its method of work |
| CN108020781A (en) * | 2017-12-19 | 2018-05-11 | 上海电机学院 | A kind of circuit breaker failure diagnostic method |
| CN109190828A (en) * | 2018-09-07 | 2019-01-11 | 苏州大学 | Gas leakage concentration distribution determines method, apparatus, equipment and readable storage medium storing program for executing |
| CN110689324A (en) * | 2019-09-29 | 2020-01-14 | 国网湖北省电力有限公司电力科学研究院 | Auxiliary judging method for detection result of distribution transformer |
| CN111612128A (en) * | 2020-05-14 | 2020-09-01 | 国网江苏省电力有限公司电力科学研究院 | Active early warning method and device for SF6 high-voltage circuit breaker state |
Non-Patent Citations (5)
| Title |
|---|
| Intelligent Pipeline Leak Detection and Analysis System;Huiwen Lin 等;2020 15th International Conference on Computer Science & Education (ICCSE);全文 * |
| SF_6泄漏类缺陷无线监测与预警装置;金海望 等;电气技术(第03期);全文 * |
| 基于OCSVM的气体绝缘组合电器故障检测;高玉峰;;电工电气(第05期);全文 * |
| 基于动态核独立元统计量的石油管道泄漏检测;辜小花 等;仪器仪表学报;全文 * |
| 采用EWT和OCSVM的高压断路器机械故障诊断;黄南天 等;仪器仪表学报(第12期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113945329A (en) | 2022-01-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117093879B (en) | Intelligent operation management method and system for data center | |
| CN108564254B (en) | Power distribution equipment state visualization platform based on big data | |
| EP1982301B1 (en) | Method of condition monitoring | |
| CN112611936A (en) | Distribution network transformer fault dynamic detection and classification system based on edge calculation | |
| CN114841396B (en) | Method for predicting metamorphic trend and warning catastrophe risk in petrochemical production process | |
| CN117408514A (en) | Intelligent operation and maintenance transformer substation monitoring and early warning system and method based on multi-parameter sensor | |
| CN117150283B (en) | Emergency environmental event safety early warning method based on big data analysis | |
| CN117434372B (en) | Electromagnetic compatibility immunity test method and system for electronic product | |
| CN112598144A (en) | CNN-LSTM burst fault early warning method based on correlation analysis | |
| CN112149877A (en) | Multi-source data-driven fault prediction method and system for multi-element complex urban power grid | |
| CN113255783A (en) | Sensor fault detection method and device based on unsupervised learning | |
| CN115238831A (en) | Fault prediction method, apparatus, computer device, storage medium and program product | |
| CN116861369A (en) | Air compressor health assessment method and system for industrial internet heterogeneous data fusion | |
| CN113945329B (en) | SF (sulfur hexafluoride) 6 Method and system for judging gas leakage defect | |
| CN111861141B (en) | Power distribution network reliability assessment method based on fuzzy fault rate prediction | |
| CN116863664A (en) | Real-time monitoring method and system for gas equipment | |
| CN116311739A (en) | Multi-sensor fire detection method based on long-short-term memory network and environment information fusion | |
| Jiafeng et al. | Transformer Fault Prediction Method Based on Multiple Linear Regression | |
| KR102350635B1 (en) | Apparatus and method of plant failure prediction | |
| Eklund | Using synthetic data to train an accurate real-world fault detection system | |
| CN118445736B (en) | Switch cabinet health state monitoring method and equipment | |
| Rigamonti et al. | AN ADVANCED EARLY FAULT DETECTION TOOL FOR PREDICTIVE MAINTENANCE OF A FLEET OF INDUSTRIAL GAS TURBINES | |
| CN118094278B (en) | Data quality inspection method, device and medium based on power application scene difference | |
| CN115330033A (en) | SF (sulfur hexafluoride) 6 Air chamber leakage defect detection method | |
| Aduroja | Intelligent Predictive Maintenance for Fire Detection System with Improved Holt-Winters Method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |